Doing Computational Social Science

John McLevey is an Associate Professor in the Department of Knowledge Integration at the University of Waterloo (ON, Canada). He is also appointed to the Departments of Sociology & Legal Studies and Geography and Environmental Management, is a Policy Fellow at the Balsillie School of International Affairs, and a Member of the Cybersecurity and Privacy Institute at the University of Waterloo. His work is funded by research grants from the Social Science and Humanities Research Council of Canada (SSHRC) and an Early Researcher Award from the Ontario Ministry of Research and Innovation. His current research project focuses on disinformation, censorship, and political deliberation in the public sphere across a wide variety of national contexts and political regimes. He wrote Doing Computational Social Science (SAGE Publishing, 2021) from his experiences as a researcher and advisor, as well as teaching courses in computational social science, data science, and research methods to students from diverse disciplinary backgrounds at the undergraduate and graduate level.

This book adds two key components critical for social science students that are often lacking in other texts. The first is a discussion of key computing components starting at the most basic – often these steps are overlooked in texts written for computer science students. Second is the excellent grounding in and integration with social science theory and concepts.

• Introduction: Learning to do computational social sciencePart I: FoundationsChapter 1: Setting up your open source scientific computing environmentChapter 2: Python programming: The basicsChapter 3: Python programming: Data structures, functions and filesChapter 4: Collecting data from Application Programming Interfaces (APIs)Chapter 5: Collecting data from the web: ScrapingChapter 6: Processing structured dataChapter 7: Visualisation and exploratory data analysisChapter 8: Latent factors and componentsPart II: Fundamentals of text analysisChapter 9: Processing natural language dataChapter 10: Iterative text analysisChapter 11: Exploratory text analysisChapter 12: Text similarity and latent semantic spacePart III: Fundamentals of network analysisChapter 13: Social networks and relational thinkingChapter 14: Connection and clustering in social networksChapter 15: Influence, inequality and power in social networksChapter 16: Going viral: Modelling the epidemic spread of simple contagionsChapter 17: Not so fast: Modelling the diffusion of complex contagionsPart IV: Research ethics and machine learningChapter 18: Research ethics, politics and practicesChapter 19: Machine learning: Symbolic and connectionistChapter 20: Supervised learning with regression and cross-validationChapter 21: Supervised learning with tree-based modelsChapter 22: Neural networks and deep learningChapter 23: Developing neural network models with Keras and TensorflowPart V: Bayesian machine learning and probabilistic programmingChapter 24: Statistical machine learning and generative modelsChapter 25: Probability: A primerChapter 26: Approximate posterior inference with stochastic sampling and MCMCPart VI: Bayesian data analysis and latent variable modelling with relational and text dataChapter 27: Bayesian regression models with probabilistic programmingChapter 28: Bayesian hierarchical regression modellingChapter 29: Variational Bayes and the craft of generative topic modellingChapter 30: Generative network analysis with Bayesian stochastic blockmodelsPart VII: Embeddings, transformer models and named entity recognitionChapter 31: Can we model meaning?: Contextual representation and neural word embeddingsChapter 32: Named entity recognition, transfer learning and transformer models